test(nlp): add case to ensure Catalan middle dot is preserved in audio text normalization

Add unit test verifying that U+00B7 middle dot is not stripped from Catalan
words during audio text preprocessing, reflecting multilingual requirements.

tests/unit/nlp.vitest.spec.ts

@@ -41,6 +41,12 @@ describe('preprocessSentenceForAudio', () => {
         input: '• Bullet ‣ marker ◦ list',
         expected: 'Bullet marker list',
       },
+      {
+        // U+00B7 middle dot is a real letter in Catalan ("col·legi") and must
+        // NOT be stripped like decorative bullets, since OpenReader is multilingual.
+        input: 'Col·legi de l·l',
+        expected: 'Col·legi de l·l',
+      },
       {
         input: 'Multiple    spaces',
         expected: 'Multiple spaces',